An automated sequential injection (SI) on-line solvent extraction-back extraction eparation/preconcentration procedure is described. Demonstrated for the assay of cadmium by electrothermal atomic absorption spectrometry (ETAAS), the analyte is initially complexed with ammonium pyrrolidinedithiocarbamate (APDC) in citrate buffer and the chelate is extracted into isobutyl methyl ketone (IBMK), which is separated from the aqueous phase by means of a newly designed dual-conical gravitational phase separator. A metered amount of the organic eluate is aspirated and stored in the PTFE holding coil (HC) of the SI-system. Afterwards, it is dispensed and mixed with an aqueous back extractant of dilute nitric acid containing Hg(II) ions as stripping agent, thereby facilitating a rapid metal-exchange reaction with the APDC ligand and transfer of the Cd into the aqueous phase. The aqueous phase is separated in a second dual-conical gravitational phase separator, and 30 l of it is entrapped and metered in a sample loop (SL) and subsequently introduced via air segmentation into the graphite tube for analyte quantification. The ETAAS determination is performed in parallel with the separation/preconcentration process of the ensuing sample. An enrichment factor of 21.4, a detection limit of 2.7ng l−1, along with a sampling frequency of 13h−1 were obtained at a sample flow rate of 6.0ml min−1. The precision (R. S. D.) at the 0.4g l−1 level was 1.8% as compared to 3.2% when quantifying the organic extractant directly. The applicability of the procedure is demonstrated for the determination of trace levels of cadmium in three certified reference materials.

Termed the third generation of flow-injection analysis, sequential injection (SI)-lab-on-valve (LOV) has specific advantages and allows novel, unique applications-not least as a versatile front end to a variety of detection techniques. This review presents and discusses progress to date of the SI-LOV approach as well as its applications in the automation and the micro-miniaturization of on-line sample pre-treatment. Special emphasis is placed on using SI-LOV in conjunction with bead injection (BI) for on-line separation and pre-concentration of ultra-trace levels of metals by exploiting the renewable micro-column approach. With detection by ETAAS and ICP-MS, it is shown, as illustrated by recent results in the authors' laboratory, that this methodology eliminates the problems encountered in conventional on-line column pre-concentration systems, improves the overall operational efficiency and yields the robustness necessary for routine assays. Also discussed is the future potential of the SI-LOV approach as a front end to various analytical protocols.

An automated sequential injection on-line preconcentration procedure for trace metals by using a PTFE bead-packed microcolumn coupled to ICP-MS is described, and used for simultaneous analyses of cadmium and lead. In dilute nitric acid (0.5%, v/v), neutral complexes between the analytes and chelating reagent, diethyldithiophosphate (DDPA), are formed and adsorbed onto the surface of the PTFE beads. The adsorbed complexes are afterwards eluted with 20% nitric acid and the leading part of the eluate (40ml) is stored in a sample loop (SL), the contents of which are subsequently transported, via a direct injection high efficiency nebulizer (DIHEN), into the ICP-MS for quantification. The packed column (PC) generates considerably lower hydrodynamic impedance than other commonly used sorbent columns, while its preconcentration efficiencies, in terms of retention efficiency and enrichment factor, are significantly improved as compared to a knotted reactor (KR). Inherently, the column approach results in low blank values and consequently in low limits of detection (LODs). With a 40s sample loading time at 4.5ml min21, along with a sampling frequency of 18h21, the retention efficiencies/enrichment factors for cadmium and lead were 96%/72 and 104%/81, respectively, versus 45%/34 and 54%/41 for a knotted reactor with similar internal surface area. In addition, the limits of detection (LODs) and precisions were at the same levels, i.e., the LODs were 2.9ng l21 Cd (PC), 3.5ng l21 Cd (KR), 6.0ng l21 Pb (PC) and 4.7ng 121 Pb (KR), while the RSDs were 1.9% (Cd, PC), 2.7% (Cd, KR), 2.2% (Pb, PC) and 2.5% (Pb, KR). The procedure was validated by determination of trace cadmium and lead in certified reference materials and urine samples.

% for the procedure in which the loaded beads are transported directly to the graphite furnace for pyrolysis and atomization, and even improved in comparison to the traditional unidirectional and bidirectional repetitive elution procedures which under comparable conditions yield R.S.D.-values of 5.8 and 4.9%, respectively. The tolerance limits for cations such as Pb (II), Zn (II), Co (II) and Mn (II) were improved up to 10-50-folds, and the linear calibration range extended to comprise 0.02-1.20g l−1. Because of lower operating temperature, the life time of the graphite tube is extended. An enrichment factor of 71.1 and a detection limit of 10.2ng l−1 along with a sampling frequency of 12 h−1 were obtained, which are at the same levels as those for the previously described procedure without elution. The present approach was validated by determination of the nickel contents in two certified reference materials, an industrial waste water sample and a human urine sample.

A sequential injection (SI) on-line lab-on-valve separation/preconcentration system incorporating a renewable ion-exchange micro-column interfaced with inductively coupled plasma mass spectrometry (ICP-MS) via a home-made direct injection high efficiency nebulizer (DIHEN) is described. Aimed at eliminating spectroscopic and non-spectroscopic interferences, the renewable micro-column is first loaded by aspirating 15ml of a SP Sephadex C-25 cation-exchanger bead suspension and then exposed to a defined volume of sample solution. Residuals of matrix components are pre-eluted with carrier solution (nitric acid,1: 80v/v), and the measurands are subsequently eluted with a defined volume of nitric acid (1: 16v/v). The leading volume of eluate (ca. 30ml) is collected and introduced into the plasma via the DIHEN. The eluted beads are then discarded and a new column is aspirated for the next operation. Data acquisition is performed in parallel with the ensuing preconcentration process. With 2.0ml sample loading, the enrichment factors are 35.2 (0.05-2.4mg l21) for Ni, and 9.1 (0.04-1.6mg l21) and 28.4 (1.6-3.2mg l21) for Bi. The detection limits (3s) are 15 ng l21 (Ni) and 4 ng l21 (Bi), while the sampling frequency is 12 per hour. The precisions are 2.9% (0.8mg l21 Ni) and 1.7% (0.8mg l21 Bi). The procedure is validated by determination of nickel and bismuth in a certified reference material (CRM 320, River Sediment), and the recoveries are measured by spiking two human urine samples.